Exploratory and Confirmatory Factor Analysis of PERMA for ...
Exploratory and Confirmatory Factor Analysis Study …...Exploratory and Confirmatory Factor...
Transcript of Exploratory and Confirmatory Factor Analysis Study …...Exploratory and Confirmatory Factor...
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
44
Exploratory and Confirmatory Factor Analysis Study for the
Scale of Attitudes towards Chemistry Homework
Ayşem Seda Önen
Assoc. Prof. Dr., Department of Chemistry Education, Hacettepe University, Ankara, Turkey
[email protected], Phone: +90 312 297 67 83 Fax: +90 312 297 8600
Canan Koçak
Dr., Department of Chemistry Education, Hacettepe University, Ankara, Turkey
Abstract
In this study, new items were added to the “Scale of Attitudes Towards
Homework”, which was developed by Yucel (2004) with the aim of determining
secondary school students’ attitudes towards chemistry homework tasks they are
given. The final version of the scale was also analyzed in terms of its validity and
reliability via the exploratory and confirmatory factory analysis. For the reformation
of the assessment tool, the sampling (n=250) was enhanced up to 630 secondary
school students. Following steps were taken in the development of the Scale of
Attitudes Towards Chemistry Homework Tasks: (1) Literature review and
construction of the item pool, (2) Taking expert opinions, (3) Total correlations of
items, (4) Item distinctiveness, (5) Exploratory factor analysis, (6) Guttman Spithalf
coefficient internal consistency reliability obtained through Cronbach Alpha and
Split-half method. (7) Examination of the correlations between subscales (8)
Confirmatory factor analysis.The Cronbach Alpha reliability coefficient of the scale
was calculated to be .85 and the conclusions of the confirmatory factor analysis
supported the structure with three subdimensions. Confirmatory factor analysis
concluded that the model had an acceptable fit value and the scale was reliable and
valid in terms of determining positive and negative attitudes of students towards
chemistry homework tasks. In other words, the final version of the scale was found
to be reliable and valid in terms of determining the attitudes of secondary school
students towards homework.
Keywords
The scale of attitudes towards chemistry homework tasks, reliability, validity,
exploratory factor analysis, confirmatory factor analysis.
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
45
Introduction
Academic achievement levels of students have been accepted as an important indicator of
the education system in Turkey as well as many countries. With the aim of determining the
failures and elements in need of improvement within the education system, homework has
been a common topic of evaluation activities. Homework is a task given to students to
perform outside the school for enhancing and enabling the permanence of the learnt
knowledge or experiences. Students take the advantage of homework studies by reinforcing
their knowledge when revising or solving problems through focusing on the points they
miss during the lessons (Sarigöz, 2011). Homework, as one of the most efficient learning
activity outside the school, is a functional tool when students lack the adequate time for
individual work or practice in the classroom due to high number of students in a classroom
or lack of time allocated (Yucel, 2004). Students, who complete their homework tasks timely,
not only revise what they learn during the lesson but also develop their own ways of
studying for reviewing the topics (Paulu & Perkinson, 1995). Homework tasks are not
necessarily written activities. Students should be able to use new tools and perform new
experiments. Hence, further observation and experiments facilitate students in becoming
aware of their knowledge and skills while affecting their development on the positive (Bryan
& Burstein, 2004; Cooper, 2001; Macbet, 2003; Swanson, 2001).
Constructivist approach, as one of the latest trends in education, has affected the visions of
many countries by its focus on the knowledge, construction of the knowledge and its usage
(Acat, Anılan & Anagun, 2007). It has been embraced in Turkey starting from the 2004-2005
academic year. Major changes in educational approaches and practices require the reflection
of these changes in the homework tasks given to students. Students’ roles in the
constructivist approach are to research, examine, evaluate and learn on their own while
teachers take the role of guides as experts. This causes the constructivist approach to be
considered as an approach involving irregularity and lacking order. However, constructivist
approach requires more order in the classroom than traditional approaches, which may
sometimes be complex for the inexperienced teachers (Foote, Brophy & Battaglia, 2001;
Karacaoğlu, 2010). Similarly, homework tasks assigned by inexperienced teachers remain
meaningless and dysfunctional for students. Constructivist approach requires the
assignment of homework tasks, which enable students to transfer knowledge into new
situations, make comparisons, share knowledge and aim further than simply accomplishing
activities and presenting them in the classroom (Cooper, Lindsay, Nye & Greathouse,1998;
Cooper, 2001; Corno, 2000; Gill & Shlossman, 1996).
Homework assignments do not only provide the revision of the topics. They serve various
aims such as decreasing the possibility of forgetting, reinforcing learnt topics, keeping
students’ minds active, developing research and expression skills, enabling students to
organize their ideas, teaching students to reach the sources of knowledge, ensuring the
attainment of problem solving skills, making cause and effect connections and improving
self confidence levels (Cooper, 2001; Doyle & Barbar, 1990; Fehrman, Keith, & Reimers,
1987; Foyle &Bailey, 1985; Gur, 2002; Hong, 1998; Kathleen, 2001; Krapp, 1988; Kucukahmet,
2004; Neilson, 2005; Okutan, 1990; Onal, 1995; Smith, 1989; Tan, 2005; Temel 1989; Yesilyurt,
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
46
2006). It is very important to assess to what extent these goals are achieved as well as the
effectiveness of the practice at all levels (Aladag & Dogu, 2009; Batan, 2007; Buyuktokatli,
2009; Cahit & Dogan, 2009; Cetinkaya, 1992; Gur, 2002; Kapikiran & Kiran, 1999; Kiciroglu,
2004; Ocal, 2009; Sarıgoz, 2011; Yesilyurt, 2006; Yucel, 2004; Yucel, 2008). In the light of this
view, the “Scale of Attitudes Towards Homework”, which was developed by Yucel (2004)
with the aim of determining attitudes of secondary school students towards chemistry
homework tasks was improved. In order to increase the reliability of the data obtained from
the scale, the number of items were increased by addition of items, which are thought to
exemplify the qualities to be assessed and distinguish between sub-categories. The sampling
group (n=250) was extended (n=630). The reliability and validity studies were repeated
through exploratory and confirmatory factor analysis. It is assumed that the assessed data
shall depend on a single treat as much as possible. However, considering the content of each
item, it is difficult to say that it assesses a single dimension (Ozguven, 2004). Therefore, the
scale, which had a single dimension in the previous version, was analyzed for potential
explanation on various variables in meaningful, interrelated and independent factors that
are less in number (Cronbach, 1990; Hair, Black, Babin, Anderson & Tatham, 2006; Nunnally,
1967). Acording to Cokluk, Sekercioğlu and Buyukozturk (2010) knowledge about the
structure of assessment could increase with the exploratory factor analysis. However, in
order to obtain detailed knowledge, this analysis could remain inadequate. Therefore, many
studies involve both exploratory and confirmatory factor analysis as they have known and
unknown variables. In order to display the factor pattern of the assessment tool, it is
required that confirmatory analysis shall be administered along with the exploratory
analysis. Hence, the new version of the scale was applied confirmatory factor analysis in
order to assess whether it supported the expected structure or not.
Methods
This study aimed to develop a Likert-type scale by restructuring the “Scale of Attitudes
Towards Homework “developed by Yucel (2004) with the aim of determining secondary
school students’ attitudes towards chemistry homework tasks. To achieve this aim, a total of
630 (487 female and 143 male) students studying at 9th, 10th, 11th and 12th grades of general
secondary schools were selected as sampling. As Tabachnick and Fidell defines 300
participants as “good”, 500 as “very good”, and “1000” as “perfect” number of sampling in
factor analysis, the pilot study concluded that adequate data were obtained for the statistical
analysis.
Results of Research
Preparation of the Data Collection Tool
Since it is impossible to observe behaviors of individuals directly, attitudes could be derived
from their observable behaviors (Tavşancil, 2010). The “Scale of Attitudes Towards
Chemistry Homework Tasks” was developed in the light of this assumption. Following
steps were taken in the development of the Scale of Attitudes Towards Chemistry
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
47
Homework Tasks: (1) Literature review and construction of the item pool, (2) Taking expert
opinions, (3) Total correlations of items, (4) Item distinctiveness, (5) Exploratory factor
analysis, (6) Guttman Spithalf coefficient internal consistency reliability obtained through
Cronbach Alpha and Split-half method. (7) Examination of the correlations between
subscales (8) Confirmatory factor analysis.
Review and Item Pool Construction
The most important challenge in developing an original scale is writing items appropriate to
the theoretical structures of characteristics to be assessed. Therefore, the theoretical
structures of characteristics shall be well-examined and researched in terms of dimensions.
Otherwise, obtained data may fail to support the theoretical structures of the characteristic
(Seker & Gencdogan, 2006). Therefore, during the preparation of the Scale of Attitudes
Towards Chemistry Homework Tasks the theoretical structure was carefully examined with
the aim of constructing the scale format (Cooper, 2001; Corno, 2000; Cowan, & Hallam, 1999;
Kralovec & Buell, 2001; Mengel, Holchroft & Okan, 1989; Zahn, 1966); similar scales with
different sampling were searched (Berberoglu, 1990; Gramling, 2011; Sarigoz, 2011; Xu, 2007)
and statistical resources for developing a data collection tool were analyzed (Bindak, 2005;
Cokluk, Sekercioğlu & Buyukozturk, 2010; Cronbach, 1990; Hair, et al., 2006; Kalayci, 2006;
Nunnally, 1967; Ozguven, 2004; Tabachnick & Fidell, 1996, 2001). Then, the obtained
statements were added to the scale previously developed by Yucel (2004). More than a single
item was considered for each attitude to be questioned. Moreover, attitude items were
reorganized in terms of sentence length and Turkish grammatical structure. An item pool of
30 items, which were thought to be adequate for distinguishing and to exemplify the
attitudes to be assed, was created. The direction of an attitude means its emotional
characteristic such as displaying likes or dislikes as well as positive and negative feelings.
The degree of an attitude reflects the acceptance or refusal according to the level of its
direction (Ozguven, 2004). Therefore, it was decided that the Scale of Attitudes Towards
Chemistry Homework Tasks shall be in 5-point Likert-type, which is common in the field of
social sciences.
Taking Expert Opinions
For providing the content validity of the Scale of Attitudes Towards Chemistry Homework
Tasks, expert opinions were taken. Experts examined the scale in terms of its capacity to
assess students’ attitudes towards chemistry homework tasks successfully as well as its
appropriateness to the age levels of secondary school students. The number of items was
reduced according to the expert opinions and a total of 26 items were obtained with 14
positive and 12 negative attitude contents.
Application Stage
Legal permissions were taken from the Ankara City Directorate of Education for applying
the study in the certain secondary schools. Randomly chosen 630 students formed the
sampling of the pilot study. The data obtained were analyzed in terms of students’ responses
to all items or a single one as a means of pre selection. As a result of this stage, 607 data were
uploaded to computers for evaluation purposes regarding the analysis. Since the analyses
would be done via SPSS 15.0 and Lisrel 8.7, the positive items of the scale were coded from 5
to 1 indicating “Strongly agree” to “Strongly disagree”. Negative items were coded from 1 to
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
48
5 on the opposite way. Therefore, the highest score that could be obtained from the trial
format was 130 and the lowest was 26. The statistical results regarding the reliability and
validity analysis of the scale are explained in the following related sections of this paper.
Item Analysis
Item analysis is one of the statistical analyses made in scale development studies. Item
analysis is made with the aim of removing or revising the ineffective items in a data
collection tool (Ozguven, 2004) and it presents strong data on the structural validity of the
scale. Therefore, item-total correlation-based item analysis was made for the validity and
reliability studies of the Scale of Attitudes Towards Chemistry Homework Tasks along with
the independent group-t test made for the comparison of 27% sub group, main group and
item scores. Students’ scores were listed from highest to lowest and top 163 students, which
formed the 27% of the sampling, were coded as main group (1), while the last 163 students
were coded as sub-group (2). The values obtained from the evaluation of the scale are
displayed on Table 1.
Table 1. Item analysis values of scale of attitudes towards chemistry homework tasks.
1: Main
2: Sub t*
Item total
Correlation
1:Main
2: Sub
t* Item total
Correlation
1 1 3.75 10 .41 14 1 3.92 9.1 .33
2 2.32 2 2.69
2 1 3.88 12 .45 15 1 3.62 8.4 .37
2 2.34 2 2.47
3 1 3.00 3 .10 16 1 3.43 6.9 .31
2 2.51 2 2.50
4 1 4.18 12 .47 17 1 3.84 12 .44
2 2.67 2 2.34
5 1 4.12 13 .49 18 1 3.96 12 .47
2 2.53 2 2.51
6 1 3.56 8 .31 19 1 3.99 12 .46
2 2.45 2 2.47
7 1 4.01 11. .46 20 1 3.61 8.3 .35
2 2.60 2 2.55
8
1
2
3.55
5 .24 21
1 3.89
2.49
11 .44
2 2.75 2
9 1 4.11 14. .48 22 1 3.49 7.5 .26
2 2.42 2 2.44
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
49
*p<.01
As Table 1 displays, the t-value of the difference between the attitude score averages of main
and sub groups for each item was calculated and significant differences between the groups
were examined. Moreover, the item-total correlation coefficient values were examined and
found to change between .10 and .49. In item analysis, in order to protect the summability
aspect of the scale, it has to be higher than .25 (Ozdamar, 1997). According to the analysis
results, the item total correlations of items 3, 8 and 26 had values lower than .25. Therefore,
these items were removed from the scale as they destroyed the internal consistency of the
scale. The validity and reliability degree of the scale depends mainly on the quality and
qualifications of its items. Therefore, good items were chosen through item analysis and the
reliability and validity of the data collection tool were ensured (Ozguven, 2004).
Exploratory Factor Analysis
In order to obtain proof for the structural validity of the Scale of Attitudes Towards
Chemistry Homework Tasks, exploratory factor analysis was applied. This factor analysis is
used when turning various interrelated variables into meaningful and independent factors,
which are less in number (Hair, et al., 2006). However, in order to decrease the possible
problems, the size of the sampling and normality require to be questioned before the
exploratory factor analysis studies (Cokluk, Sekercioglu & Buyukozturk, 2010). Therefore,
the appropriateness of the data structure for the factor analysis in terms of sampling size was
assessed through Kaiser-Meyer-Olkin measure of sampling adequacy test. The normal
distribution of the data was tested with the Barlett globalization test values.
Table 2. Scale of attitudes towards chemistry homework tasks and barlett test analysis results.
Kaiser Meyer Olkin Measure of Sampling Adequacy .849
Barlett Globalization Test χ2 3913.0
df 210
p .000
Table 2 displays that the Kaiser-Meyer-Olkin sampling value was calculated to be 0.84 and
the Barlett Globalization Test result was found to be significant [χ2 = 3913; p< .01]. Kaiser-
This is described as “good” in the literature (Hair et al, 2006; Tavsancil, 2010) and the
10 1 4.07 13. .49 23 1 3.77 10 .43
2 2.40 2 2.46
11 1 4.19 10. .41 24 1 3.61 8.1 .36
2 2.87 2 2.50
12 1 4.25 14. .49 25 1 3.61 9.4 .35
2 2.54 2 2.33
13 1 4.02 10 .44 26 1 3.01 3.1 .11
2 2.68 2 2.53
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
50
appropriateness of the data structure were proven. Principal components analysis was
applied as the most frequently used factorization techniques of exploratory factor analysis
(Tabachnick & Fidel, 2001). The Varimax Vertical rotation technique was used for rotation
and 3 factors were found to have eigenvalues () higher than 1.00. Later, the overlapping of
the items and their factor loading values were evaluated according to the literature (Cokluk,
Sekercioglu & Buyukozturk, 2010; Hair et al., 2006; Tabachnick & Fidel, 2001) and the items
with factor loading values equal to 0.40 or higher, and the items with more than .01
differences in their both loading values were determined. Therefore 21 items with factor
loading values between. 46 and. 83 were included in the final version of the scale while items
7 and 17 were removed from the scale as they could not match the literature scope.
21191715131197531
6
5
4
3
2
1
0
Figure 1. Scree plot for scale of attitudes towards chemistry homework tasks
According to the Scree Plot, which serves to the major aim of the factor analysis and helps in
determining the number of factors according to its contributions to the variance (Cokluk,
Sekercioglu & Buyukozturk, 2010), the point where the graphic curve has an instant fall is
where the third factor is located. Therefore, the items of the Scale of Attitudes Towards
Chemistry Homework Tasks were decided to be listed under three factors. One of these
factors was called “Usefulness” with its content of benefits of chemistry homework tasks.
The second factor with negative attitudes was named as “Negativity” and the third factor
with items containing positive attitudes was called “Positivity”. Some values obtained from
the exploratory factor analysis are displayed on Table 3.
Table 3. Exploratory factor analysis on the scale of attitudes towards chemistry homework tasks.
Factors
Fac
tor
Lo
ad
ing
Des
crib
ed
per
cen
tag
e o
f
dif
fere
nce
s
Av
erag
e
F
α
p
Factor 1 (Usefulness)
5.3 20.2 3.36 9.04 .84 .0001
11 .759
12 .690
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
51
9 .689
10 .685
19 .620
13 .601
21 .601
4 .596
5 .526
18 .516
Factor 2 (Negativity)
2.6 13.9 3.0 7.07 .74 .0001
15 .725
16 .684
20 .613
6 .594
24 .579
25 .530
14 .525
22 .464
Factor 3 (Positivity)
1.4 11.0 3.1 7.0 .72 .0001
1 .831
2 .819
23 .515
Kaiser-Meyer-Olkin Measure of Sampling= 0.84; The (variance) explanation rate of the total difference: 45.315
Table 3 displays that the factor loading of items grouped under the structure with three
factors had higher values than acceptable and the variance explanation score of the three
factors was found to be 45.31%. This value is described as adequate in social sciences (Scerer
et.al., 1988; adapted by Tavsancil, 2005). Similarly, the contribution of the factors to the total
variance was 20.2% for the first factor, 13.9% for the second factor and 11% for the third
factor, which is adequate. These findings show that the Scale of Attitudes Towards
Chemistry Homework Tasks has a strong factor structure. In order to assess whether the
three sub-dimensions of the Scale of Attitudes Towards Chemistry Homework Tasks
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
52
addressed the same characteristics, Pearson Multiplication of Moments was calculated
among the three factors. The findings are displayed on Table 4.
Table 4. The correlation coefficients of scale of attitudes towards chemistry homework tasks factors.
Factor Usefulness Negativitiy Positivitiy
Usefulness 1 .21 .52
Negativitiy .21 1 .22
Positivitiy .52 .22 1
p<.01
Table 4 shows that there are positive significant relationships between the factors of the
scale. In other words, factor variables were found to be supportive of each other (r = .52, r =
.21, r = .22, p = 0,0001).
The Reliability Study of the Scale of Attitudes Towards Chemistry Homework Tasks
For the reliability analysis of the Scale of Attitudes Towards Chemistry Homework Tasks,
Cronbach Alpha Coefficient and Split Half reliability analysis was administered to the whole
scale and all sub dimensions. The statistical analysis concluded that the Cronbach Alpha (α)
was. 85 and the Guttman Splithalf coefficient determined via the Split-half method was 0.84.
This is another finding that the scale is reliable (Nunnally, 1967). Additionally, The Cronbach
Alpha Reliability Coefficient was calculated for the first sub-dimension as .84, for the second
sub-dimension as .74 and for the third sub-dimension as .72. This shows that all items of the
Scale of Attitudes Towards Chemistry Homework Tasks are consistent. Shortly, the
reliability analysis resulted that the Scale of Attitudes Towards Chemistry Homework Tasks
was reliable enough to be accepted in research on the social sciences.
Confirmatory Factor Analysis
Following the exploratory factor analysis, confirmatory factor analysis was made for
questioning the validity of the obtained structure. For displaying the factor pattern of the
assessment tool, it is required to analyze the assessment tool through confirmatory
techniques (Cokluk, Sekercioglu & Buyukozturk, 2010). The 3-dimensional structure of the
scale was tested for confirmation through the structural equality modeling technique. The
Path Diagram obtained from the Lisrel 8.7 software is displayed on Figure 2.
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
53
Figure 2. Path diagram for scale of attitudes towards chemistry homework tasks.
The model was questioned based on the literature in terms of its production of acceptable
values (Brown, 2006; Byne, 1994; Cokluk, Sekercioglu & Buyukozturk, 2010; Hair, et al., 2006;
Hooper, Caughlan & Muller, 2008; Joreskog & Sorbom, 1993; Schumacker & Lomax, 1996;
Sumer, 2000; Tabachnick & Fidell, 2001; Kline, 2005; Raykov & Marcoulides, 2008; Yilmaz &
Celik, 2009), and the results were shown on Table 5.
Table 5. Fit measures of the model
The model was found to have acceptable fit values as a result of the confirmatory factor
analysis as displayed on Table 5. In other words, the confirmatory factor analysis results
supported the finding that the scale has three dimensions. These findings show that the scale
is a reliable and valid data collection tool in determining students’ attitudes towards
chemistry homework tasks.
Results of Research
Many variables could be assessed in education. However, this requires assessment tools,
which facilitate the observation of the size to be assessed and its mathematical expression.
Each tool enables the observation to be made more sensitively. Assessment tools used in
education vary according to the structure of the size to be assessed and the group that the
Fit Measure Value Fitting
χ2/df
3.0 Good
RMSEA .060 Good
GFI .090 Good
AGFI .087 Acceptable
RMR 0.05 Good
NFI 0.91 Good
NNFI 0.93 Good
CFI 0.94 Good
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
54
tool shall be administered to. For assessment, the assessment tools appropriate to the goal of
the evaluation shall be selected or developed (Turgut, 1995). In the light of this view, the
“Scale of Attitudes Towards Homework”, which was developed by Yucel (2004) with the
aim of determining attitudes of secondary school students towards chemistry homework
tasks was improved with the addition of new items2.
The reliability and validity were analyzed for the final version of the study. In construction
and changing of attitudes, educational institutions and teachers have important effects
(Tavşancil, 2010). However, while questioning these effects, students’ attitudes shall also be
determined. Therefore, various assessment tools have been developed and applied with the
aim of assessing students’ attitudes towards chemistry homework tasks (Berberoglu, 1990;
Gramling, 2011; Gurlevik, 2006; Oguz, 2010; Sarigoz, 2011; Tilgel, 2010; Xu, 2007; Yucel,
2004). In this study, the scale that was developed by Yucel (2004) was restructured to serve
better to analyzing students’ attitudes towards chemistry homework tasks from different
perspectives. For displaying the factor pattern of the tool, both exploratory and confirmatory
factor analyses were applied. The process of improving the Scale of Attitudes Towards
Chemistry Homework Tasks was initiated with the literature review and item pool
construction. The previously prepared pool of 45 items was reevaluated and an item pool of
30 items was obtained. After taking the expert opinions, the number of items decreased to
26. A pilot study was made with the participation of 630 secondary school students. The data
were evaluated and 607 data were found to be appropriate to analysis. Statistical analysis
started with item analysis. Item-total correlation-based item analysis was made for the
validity and reliability studies of the Scale of Attitudes Towards Chemistry Homework
Tasks along with the independent group-t test made for the comparison of 27% sub group,
main group and item scores. Factor analysis stage followed the item analysis. Prior to the
exploratory factor analysis, in order to test the sampling size for appropriateness in
factorization, Kaiser- Meyer- Olkin test was applied and the result was found to be 0.84. This
showed that the sampling size was adequate for factor analysis (Hair et.al., 2006). Moreover,
the Barlett Globalization Test results concluded with a qhi-square value of significance at the
0.0001 level [χ2 = 3913.005]. Principal components analysis and Varimax Vertical Rotation
technique were used for the exploratory factor analysis. The analyses concluded with three
factors with eigenvalues () higher than 1.00. In the end, 21 items with factor loading values
between .46 and .83 were included in the scale, while the items 7 and 17 were removed as
they failed to comply with the acceptability standards. The three dimensional structure of
the Scale of Attitudes Towards Chemistry Homework Tasks was made certain with the Scree
Plot Graph. The first of these factors was called “Usefulness” with its content of benefits of
chemistry homework tasks. The second factor with negative attitudes was named as
“Negativity” and the third factor with items containing positive attitudes was called
“Positivity”.
The Scale of Attitudes Towards Chemistry Homework Tasks was found to have a strong
factor structure as their factor loadings were found to be higher than acceptable values and
the explanation rate of the three factors was adequate being 45.31%. In order to assess
2 Developed scale can be found at the corresponding author's e-mail (Contact:[email protected]).
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
55
whether the three sub-dimensions of the Scale of Attitudes Towards Chemistry Homework
Tasks addressed the same characteristics, Pearson Multiplication of Moments was calculated
among the three factors. The analysis concluded with the determination of positive
relationship among the factors (r = .52, r = .21, r = .22, p = 0,0001). This finding was accepted
as supportive of the consistent factor structure of the scale. The reliability of the scale was
analyzed through Cronbach Alpha Coefficient and Split Half, which concluded with the
Cronbach Alpha (α) as. 85 and the Guttman Splithalf coefficient determined via the Split-half
method as 0.84. This is another finding that the scale is reliable (Nunnally, 1967).
Additionally, The Cronbach Alpha Reliability Coefficient was calculated for the first sub-
dimension as. 84, for the second sub-dimension as. 74 and for the third sub-dimension as. 72.
This shows that all items of the Scale of Attitudes Towards Chemistry Homework Tasks are
consistent. Finally, the three-dimensional structure obtained was confirmed via the
structural equality modeling. Confirmatory factor analysis concluded that the model had an
acceptable fit value (χ2/df=3, RMSEA=0.06, GFI=0.90, AGFI=0.87, RMR= .050, NFI= 0.91,
NNFI= 0.93, CFI= 0.94) and the scale was reliable and valid in terms of determining positive
and negative attitudes of students towards chemistry homework tasks. However, the
administration of the Scale of Attitudes Towards Chemistry Homework Tasks could be
applied to a greater number of students from different cities of Turkey for further
contribution to both the scale and the subject field.
Acknowledgements
Along with the effects of homework tasks on secondary school students’ achievements, their
attitudes should also be determined. Therefore, this restructured scale is thought to serve the
literature as a reliable and valid assessment tool in determining students’ attitudes
Determination of the dimensions of the attitudes is thought to have various benefits for the
researchers of the subject field. With the help of the developed assessment tool, secondary
schools students’ attitudes towards their homework tasks would be able to be analyzed in
terms of their different dimensions. Determination of students’ attitudes through the scale
would set a great resource for teachers in assigning homework tasks for their students.
Therefore, determination of students’ attitudes through the scale and the analysis of the
obtained data are thought to contribute not only the literature but also the whole world of
education.
References
Acat, B., Anılan, H., & Anagün, S.S. (2007). Yapılandırmacı öğrenme ortamlarının
düzenlenmesinde karşılaşılan sorunlar ve çözüm önerileri. VI. Ulusal Sınıf Öğretmenliği
Sempozyumu, Eskisehir, 27-29 Nisan.
Aladağ, C., & Doğu, S. (2009). Fen ve teknoloji dersinde verilen ödevlerin öğrenci
görüşlerine göre değerlendirilmesi. Selçuk Üniversitesi Sosyal Bilimler Enstitüsü Dergisi,
21, 15-23.
Batan, B. (2007). İlköğretim öğrencilerinin ev ödevleri ile ilişkili tutumlarını etkileyen faktörler.
Yüksek Lisans Tezi, Ankara Üniversitesi Eğitim Bilimleri Enstitüsü, Ankara.
Berberoğlu, G. (1990). Kimyaya ilişkin tutumların ölçülmesi. Eğitim ve Bilim, 76. 16-27.
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
56
Bindak, R. (2005). Tutum ölçeklerine madde seçmede kullanılan tekniklerin
karşılaştırılması. İnönü Üniversitesi Eğitim Fakültesi Dergisi, 6 (10), 17-26.
Brown, T. A. (2006). Confirmatory factor analysis for applied research. New York: The
Guilford Press.
Bryan, T., & Burstein, K. (2004). Improving homework completion and academic
performance: Lessons from special education. Theory Into Practice, 43(3), 213–219.
Byrne, B.M. (1998). Structural equation modeling with LISREL, PRELIS, and SIMPLIS: basic
concepts, applications, and programming. New Jersey: Lawrence Erbaum Associates
Publisher.
Büyüktokatlı, N. (2009). İlköğretimde ev ödevi uygulamalarına ilişkin öğretmen görüşlerinin
incelenmesi. Yayınlanmamış Yüksek Lisans Tezi, Selçuk Üniversitesi Sosyal Bilimler
Enstitüsü, Konya.
Cahit, Y., & Doğan, S. (2009). Fen ve teknoloji dersinde öğrencilerin ev ödevi
performansını arttırmaya yönelik bir eylem araştırması. Çankaya Üniversitesi Fen
Edebiyat Fakültesi Dergisi, 12, 221-238.
Cooper, H. (2001). Homework for all in moderation. Educational Leadership, 34-38.
Cooper, H., Lindsay, J.J., Nye, B., & Greathouse, S. (1998). Relationships among attitudes
about homework, amount of homework assigned and completed, and student
achievement. Journal of Educational Psychology, 90(1), 70–83.
Corno, L. (2000). Looking at homework differently. The Elementary School Journal, 100,
529–548.
Cowan, R.S., & Hallam, S. (1999). What do we know about homework. Institute of Education
Viewpoint 9, London: University of London.
Cronbach, L. J.(1990). Essentials of psychological testing. New York, N.Y.: Harper & Row.
Çetinkaya, A. (1992). Adana ili merkez ortaokullarında ev ödevlerine ilişkin karşılaşılan
sorunlar. Yüksek Lisans Tezi, Çukurova Üniversitesi Sosyal Bilimler Enstitüsü, Adana.
Çokluk, Ö., Şekercioğlu, G., & Büyüköztürk, Ş. (2010). Sosyal bilimler için çok değişkenli
istatistik. SPSS ve LISREL uygulamaları. Ankara: Pegem Akademi.
Doyle, M. E. & Barber, B. S. (1990). Homework as a learning experiences (3rd. Edition).
Washington, D.C: National Education Association.
Fehrman, P. G., Keith, T. Z., & Reimers, T. M. (1987). Home influence on school learning:
direct and indirect effects of parental involvment on high school grades. Journal of
Educational Research, 80, 330-337.
Foote,C.J., Vermette, P.J., & Battaglia, C. (2001). Constructivist strategies: meeting standards
and engaging adolescent minds, eye on education. N.Y: Larchmont.
Foyle, H. C. & Bailey, G. D. (1985). Homework in the classroom: Can it make a difference in
student achievement?
http://eric.ed.gov/ERICDocs/data/ericdocs2sql/contentstorage01/0000019b/80/30/3f/aa.
pdf (Web Adresinden Ekim 2008 Tarihinde Alınmıştır.)
Gill, B. P., & Schlossman, S. L. (1996). A sin against childhood: Progressive education and
the crusade to abolish homework, 1897–1941. American Journal of Education, 105, 27–66.
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
57
Gramling, M. A. (2011). Effect of two intervention strategies on science homework completion. A
professional paper submitted in partial fulfillment of the requirements for the degree
of Master of Science in Science Education, Montana State University Bozeman,
Montana.
Gür, H. (2002). Ev ödevi yapma stillerinin akademik başarıya etkisi. From
Web:http://www.matder.org.tr/index.php?option=com_content&view=category&layo
ut=blog&id=8&Itemid.
Gürlevik, G. (2006). Ortaöğretim matematik derslerinde ev ödevlerine yönelik öğretmen ve
öğrenci görüşleri (ankara ili çankaya ilçesi örneği). Yüksek Lisans Tezi, Gazi Üniversitesi
Eğitim Bilimleri Enstitüsü Eğitim Programları ve Öğretimi Bilim Dalı, Ankara.
Hair, J., J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2006).
Multivariate Data Analysis (6th ed.).Upper Saddle River, NJ: Pearson Prentice Hall.
Hong, E. (1998). Homework style, homework environment, and academic achievement.
Journal of Research in Rural Education, 21(2).
Hooper D, Coughlan, J., & Mullen ,M.R. (2008). Structural equation modelling: guidelines
for determining model fit. Elective Journal Business Research Methods, 6(1): 53-60.
Jöreskog, K., & Sörbom, D. (1993). Lisrel 8: Structural Equation Modeling with the SIMPLIS
Command Language. USA: Scienticif Software International.
Kalaycı, Ş. (2006). SPSS uygulamalı çok değişkenli istatistik teknikleri. Ankara: Asil
Yayıncılık.
Kapıkıran, Ş. & Kıran, H.(1999). Ev ödevinin öğrencinin akademik başarısına etkisi. PAÜ.
Eğitim Fakültesi Dergisi, 5, 54-60.
Karacaoğlu, Ö.C. (2010). Pisa sonuçları ve yapılandırmacı yaklaşım. Eğitim Bülteni
Dergisi, 32,11–15.
Kathleen, V. (2001). Parental involvement in homework. Educational Psychologist, 36(3),
195-209 .
Kıcıroğlu, F. (2004). Ankara il merkezinde kız meslek lisesi tekstil-konfeksiyon bölümündeki
öğretmen, öğrenci ve ailelerin ev ödevlerine yaklaşımları. Yüksek Lisans Tezi, Gazi
Üniversitesi Eğitim Bilimleri Enstitüsü, Ankara.
Kline, R. B. (2005). Principles and practice of structural equation modeling. (Second Edition),
NY: Guildford Publications, Inc.
Kralovec, E., & Buell, J. (2001). End homework now. Educational Leadership, 39-42.
Krapp, J. V. (1988). Teaching research skills: a critical-thinking approach. School Library
Media Journal, 34 (5), 18–27.
Küçükahmet, L. (2004). Öğretimde planlama ve değerlendirme. Ankara: Nobel.
Macbeth, A. (2003). Homework in europe. A discussion paper prepared for the european
parents association. Unpublished Paper.
Mengel, J. A., Holcroft, J., & Zahn, R. D. (1966). Attitudes toward homework. The
Elementary School Journal, 67(1), 41-44.
Neilson, W. (2005). Homework and performance for time-constrained students. Economics
Bulletin, 9(1), 1-6.
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
58
Nunnally, J, C. (1967). Psychometric theory. New York: McGraw Hill.
Oğuz, M.A. (2010). Ortaöğretim coğrafya derslerinde ev ödevlerine yönelik öğretmen görüşleri
(Yozgat ili örneği). Yüksek Lisans Tezi, Gazi Üniversitesi Eğitim Bilimleri Enstitüsü,
Ankara.
Okan, K. (1989). Ev ödevi hazırlama teknikleri. Ankara: Seçkin Ofset.
Okutan, M. (1990). Yeni ödev yönetmeliğinin getirdiği yenilikler nelerdir? Milli Eğitim
Dergisi, 94, 73–77.
Öcal, S. (2009). İlköğretim 4. ve 5. sınıf öğrencilerinin ev ödevlerine yönelik tutumlarının
oluşmasında ailelerin ve öğretmenlerin rolü. Yüksek Lisans Tezi, Mustafa Kemal
Üniversitesi, Sosyal Bilimler Enstitüsü, Antakya.
Önal, İ. H. (1995). Ödevlerle kazandırılan değerler. Milli Eğitim Dergisi, 126, 68–69.
Özdamar, K. (1997). Paket programlar ile istatistiksel veri analizi. Eskişehir: Anadolu
Üniversitesi Yayını.
Özgüven, İ.E. (2004). Psikolojik testler. Ankara: PDREM Yayınları.
Paulu, N., & Perkinson, K. (1995). Helping your child with homework. From Web:
http://www.kidsource.com/kidsource/content/homework.html.
Raykov, T., & Marcoulides, G. A. (2006). An introduction to applied multivariate analysis
(First Edition). NJ: Lawrence Erlbaum Associates Inc.
Sarıgöz, O. (2011). Ortaöğretim öğrencilerinin kimya derslerinde verilen ev ödevleri
hakkındaki düşüncelerinin değerlendirilmesi. Electronic Journal of Vocational Colleges,
December, 80–87.
Scherer, R. F., Wiebe, F. A., Luther, D. C., & Adams J. S. (1988). Dimensionality of coping:
factor stability using the ways of coping questionnaire. Psychological Reports, 62, 763–
770.
Smith, J. B. (1989). Library media center programs for middle schools; a curriculum based
approach. Chicago: American Library Assaciation.
Schumacker, R.E., & Lomax R.G. (1996). A beginner’s guide to structural equation modeling
(First Edition). New Jersey: Lawrance Erlbaum Associates Inc.
Sümer, N. (2000).Yapısal eşitlik modelleri. Türk Psikoloji Yazıları, 3(6) , 49–74.
Swanson, B. (2001). How important is homework.(http://
“eric.ed.gov/archives/homework.html).
Şeker, H. & Gençdoğan, B. (2006). Psikolojide ve eğitimde ölçme aracı geliştirme. Ankara:
Nobel Yayın Dağıtım.
Xu, J. (2007). Middle-school homework management: more than just gender and family
involvement. Educational Psychology, 27(2), 173–189.
Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd ed.). New York:
Harper Collins.
Tabachnick, B.G. & Fidel, L.S. (2001).Using multivariate statistics. Boston: Allyn & Bacon
Inc.
Tan, Ş. (2005). Öğretimi planlama ve değerlendirme. Ankara: Pegem A Yayıncılık.
Middle Eastern & African Journal of Educational Research, Issue 6
Year 2013
59
Tavşancıl, E. (2010). Tutumların ölçülmesi ve spss ile veri analizi. Ankara: Nobel Yayınları.
Temel, A. (1989). Öğrenci ödevleri bilinçli verilmeli. Çağdaş Eğitim Dergisi, 140, 12–14.
Tilgel, H., (2010). Attitudes of university english preparatory school students towards ICT
enhanced homewok. M.A. Thesis, Gazi University Institute of Educational Sciences,
Ankara.
Yeşilyurt, S. (2006). Lise öğrencilerinin biyoloji derslerinde verilen ev ödevlerine karşı
tutumları üzerine bir çalışma. Erzincan Eğitim Fakültesi Dergisi, 8 (1), 37–53.
Yılmaz, V., & Çelik, E. Ç. (2009). LISREL ile yapısal eşitlik modellemesi -1. temel kavramlar,
uygulamalar, programlama. Ankara: Pegem Akademi.
Yücel, S. (2004). Ortaöğretim düzeyindeki öğrencilerin kimya derslerinde verilen ev
ödevlerine karşı tutumlarının incelenmesi. Gazi Eğitim Fakültesi Dergisi, 24 (1), 147–
159.
Yücel, A. (2008). İlköğretim 7. sınıf sosyal bilgiler dersinde verilen performans ödevleri hakkında
öğretmen–veli- öğrenci görüşleri (konya örneği). Yayınlanmamış Yüksek Lisans Tezi,
Selçuk Üniversitesi Sosyal Bilimler Enstitüsü, Konya.